Marc J. Goldstein Arbitrator & Mediator NYC
June 29, 2013

The Supreme Court Returns to the Playing Field of Arbitral Power to Determine Jurisdiction

June was a fertile month for arbitration jurisprudence at the Supreme Court of the United States, and most of you know already that: 1) the Court held, 8-0, that class action arbitration is OK if the arbitrator is even arguably construing the arbitration clause when ruling that the case may proceed as a class action (Oxford Health Plans, LLC v. Sutter, No. 12-135 (Jun. 10, 2013)), (2) the Court held, 5-3, that class arbitration is not OK when the agreement expressly forbids it, even if the consequence is to make pursuit of a federal statutory treble damages claim hopelessly uneconomical – because the clause does not actually prevent a claimant for pursuing the claim, but only from winning it (American Express Co. v. Italian Colors Restaurant, No. 12-133 (Jun. 20, 2013)), and (3) the Court granted certiorari in BG Group PLC v. Republic of Argentina, where the question presented in Petitioner’s application for cert. was “In disputes involving a multi-staged dispute resolution process, does a court or instead the arbitrator determine whether a precondition to arbitration has been satisfied?

It is the third of these developments that motivates today’s post. The Court by granting the writ of certiorari has agreed to review on the merits the decision in Republic of Argentina v. BG Group PLC issued by the federal appeals court in Washington. (665 F.3d 1363 (D.C. Cir. 2012)). In that decision, the three-judge appellate panel unanimously vacated a 2007 Final Award, also unanimous, which had been issued in favor of BG Group, a UK investor, against the Republic of Argentina, in an ad hoc arbitration under the UNCITRAL Rules pursuant to the UK-Argentina bilateral investment treaty.

The UK-Argentina BIT required Investors to bring their claims to the competent courts of the Host State, and not to file arbitration until the earlier of 18 months after the court filing or an unsatisfactory final adjudication. BG Group, having concluded that Argentina had made judicial recourse untenable, bypassed the Argentine courts and commenced arbitration. The Tribunal, claiming to be engaged in treaty interpretation, held that enforcement of the litigation clause in the circumstances would be an absurd and unreasonable result, and held that it had jurisdiction over BG Group’s claim. The U.S. District Court in Washington (Washington having been selected by BG Group and Argentina as the arbitral seat) confirmed the Award (on the merits, more than $185 million in damages), but the D.C. Circuit reversed. For the D.C. Circuit, the case was not difficult: Courts should review arbitrability decisions of arbitrators de novo — absent clear and unmistakable evidence of an agreement to arbitrate arbitrability, which evidence was not present — and on a de novo review the Tribunal’s purported interpretation was in clear contradiction of the treaty’s text. For the D.C. Circuit, there was no clear evidence of an agreement to arbitrate arbitrability because the treaty itself required the Investor to file a court case and the UNCITRAL Rules and their compétence-compétence provision would not become operative until the litigation pre-condition was satisfied.

The purpose of this post is not to tell you how the Supreme Court will decide the case. (Arbitration Commentaries’ crystal ball is on holiday until 8 July). The mission here is reconciliation. And here is why: The D.C. Circuit implicitly accepts the Second Circuit position that an agreement to arbitrate under rules that contain a compétence-compétence rule constitutes the required clear and unmistakable evidence of an agreement to arbitrate arbitrability. (Herein, the “Incorpration Rule”). And thus the BG Group case might, or might appear to, turn on whether the Supreme Court accepts the D.C. Circuit’s holding (so-called “temporal limitation”) that the UNCITRAL Rules did not become operative between BG Group and Argentina under the UK-Argentina BIT because there was no agreement to arbitrate until the litigation precondition was met.  And yet the widely-admired Rapporteur of the Restatement (Third) of International Arbitration Law of the United States, Professor George Bermann, has told the Supreme Court (i) as co-counsel for a recent unsuccessful certiorari applicant in a commercial arbitration case, the Government of Thailand, that the Incorporation Rule should be rejected, at least in the context of judicial review of an arbitrator’s jurisdiction ruling (cert. denial at Docket No. 12-878 (Feb. 25, 2013); Petition for Writ of Certiorari, dated Jan. 14, 2013, at; and (ii) as co-counsel in BG Group v. Argentina for an amicus group of leading practitioners and scholars supporting BG Group, that the Supreme Court should accept certiorari (mission accomplished), reverse the D.C. Circuit, and reinstate the Award. (Find all the briefs and decisions at

In a footnote to Argentina’s reply brief opposing certiorari, Professor Bermann was accused of playing both inside and outside of the Incorporation Rule sandbox. Is this so? (Answer below: No). Or can Professor Bermann’s positions be reconciled? (Answer below: Yes).  This is the reconciliation mission of this post.

One possibly satisfying way to reconcile the positions is to recognize that formation of the agreement to arbitrate in the investment treaty context is different, and that the behavioral assumptions about parties entering into arbitration agreements in commercial contracts, made by Justice Breyer in his opinion for the unanimous Court in First Options of Chicago, Inc. v. Kaplan (514 U.S. 938 (1995)) may not, probably do not, apply to an investor invoking the offer to arbitrate made by the State Parties to a BIT. (“[T]he ‘who (primarily) should decide arbitrability question [] is rather arcane. A party often might not focus upon that question or upon the significance of having arbitrators decide the scope of their own powers.” 514 U.S. at 945). First Options tells us that courts, in deciding whether the parties agreed to arbitrate a certain matter, “generally…should apply ordinary [state law] principles that govern the formation of contracts.” 514 U.S. at 944. But the equation in the BIT context is not so simple. A court or arbitral tribunal, asked to decide whether a particular dispute in arbitrable under the BIT, or whether the investor has satisfied or must satisfy any pre-conditions to arbitration stated in the BIT, must apply (i) the text of the treaty, and/or (ii) customary international law, and/or (iii) international law instruments and principles concerning treaty interpretation, notably the Vienna Convention of the Law of Treaties.

Whereas the tools of interpretation applicable to the arbitrability issue in the investment treaty context are within the special competence of highly specialized international arbitrators, not national court judges, and whereas this fact is well understood by the State Parties to the treaty, and probably also by most investors bringing BIT claims by the time they accept investment treaty arbitration by filing claims, the presumption in the treaty arbitration context arguably should be the opposition of the presumption in the commercial arbitration context, i.e. that the parties to a BIT arbitration intend to arbitrate arbitrability unless there is clear evidence that they did not so intend. Stated another way, these contextual elements associated with arbitrability decisions under investment treaties normally should provide the “clear and unmistakable evidence” required by First Options of an agreement to arbitrate arbitrability. And that evidence is entirely separate from the compétence-compétence rules in, for example, the UNCITRAL Rules. The fact that an investment treaty arbitration is to take place under those Rules is, in terms of the “clear and unmistakable evidence” of agreement to arbitrate arbitrability, at most an embellishment.

Thus, the existence of an agreement to arbitrate arbitrability under an investment treaty arguably does not at all depend on the presence of a compétence-compétence provision in the selected arbitration rules. And if that is so, then the Supreme Court could sensibly, as urged by Professor Bermann, both (i) reject the Incorporation Rule, i.e., the Second Circuit position in cases like Contec and Schneider and Chevron v. Ecuador, and Thai-Lao Lignite, but still also (ii) reject the D.C. Circuit position in BG Group, and reinstate the Final Award and its arbitral determination of arbitrability.

* * *

This is of course not the only basis to distinguish the positions of the successful certiorari applicant BG Group and the unsuccessful one, Government of Thailand. Nor is it necessarily the theme mainly invoked by BG Group. In its petition for certiorari, BG Group largely dodged differentiation of investment and commercial arbitration. The headline of the Petitioner’s application for certiorari in BG Group is that this is a “procedural arbitrability”/”gateway question” case governed by the Court’s decision in Howsam v. Dean Witter Reynolds, Inc. (537 U.S. 79 (2002)). But when the case is argued on the merits, its classification as a “procedural arbitrability” case may emerge as problematical – because in the BIT’s dispute resolution clause, the 18 month litigation requirement is arguably not merely the pre-condition to commencing arbitration but to the existence of an arbitration agreement. The different a priori assumptions made by treaty parties and investors invoking rights under those treaties, as compared to parties to commercial contracts that have arbitration clauses, may well emerge as a vital element of the Court’s decision – and should emerge even now as a full vindication of the conceptual consistency of Professor Bermann’s advocacy at the Court.

Leave a Reply